Fenghua Weng
4 papers · 2025–2025 · 4 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Bronze
🗺️
Taxonomy Completionist
(11)
Top co-authors
Research topics
Keywords
jailbreak attack
(4)
large language model
(2)
vision-language model
(2)
safety alignment
(2)
model editing
(1)
ai safety
(1)
semantic similarity
(1)
adversarial perturbation
(1)
parameter update
(1)
defense mechanism
(1)
kl divergence regularization
(1)
adversarial suffix
(1)
optimization-based attack
(1)
benchmark evaluation
(1)
token shift
(1)
direct preference optimization
(1)
multimodal learning
(1)
adversarial training
(1)
Papers
Adversary-Aware DPO: Enhancing Safety Alignment in Vision Language Models via Adversarial Training
EMNLP 2025
DELMAN: Dynamic Defense Against Large Language Model Jailbreaking with Model Editing
ACL 2025
MMJ-Bench: A Comprehensive Study on Jailbreak Attacks and Defenses for Vision Language Models
AAAI 2025
SEMINAR: SEMantic InformatioN Augmented JailbReak Attack in LLM
ACML 2025